Does Oracle supports an index that only contains a sub-set of the records in a table?
For eg - A table has 1 million records, can we have a index which contains suppose 100k records. We have used a similar logic in DMSII in Unisys. Not so sure if Oracle supports it.
I have following records in table (T)
Idtype_cdNbrDt
1A1001/1/2013
2A101Null
3B100Null
4C100Null
5A100Null
Based on the type-cd, I need to create Unique index on Nbr & Dt columns
In above record set, I want to create Unique index on type_cd=A and not on B&C (Here i expect duplictes for a combination of Nbr & Dt)
Can we create Unique index on subset of record?
The requirement details are as below,
1) The database contains static as well as transactional data for telecom domain.
2) We have to move region-wise data from one database to another.
3) There are around 10 regions.
4) The region wise data extraction from source db is based on around 40 to 50 tables. Rest of the tables contains static data & it will not change.
5) The volume is around 1 million subscribers per region.
6) The migration is required because
a) The client is upgrading its base product which uses this database
b) There is a change in structure of static tables
c) Hardware upgrade
d) The client want to start with single region on new database & rest of the regions will be operated from old database.
7) Keep execution time to very minimum.
I am thinking to have solution as below,
1) Create destination database with upgraded db structure (as mentioned in 6b)
2) Create database links to access source db in destination db.
3) Write SQL queries to fetch data from all the respective tables for a specific region
4) Write separate PL/SQL blocks for each table to fetch data from source db & insert into respective table in destination db
a) Use FOR ALL & bulk collect to improve the performance
b) Divide table data into multiple chunks & execute parallel batches (around 10) to insert the data
5) Validate pre & post counts to verify the success of migration
How to retrieve first hundred records from a table ?
FYI
---
The table size is 5 GB
The table count is 127922653
Table has 14 columns
Table is partitioned as well.
The table has 10 partitions.
I have, for example, two tables: COMPANIES and EMPLOYEES. COMPANIES has a primary key on column COMP_ID. EMPLOYEES has a foreign key EMP_COMP_ID. I always do query COMPANIES with WHERE clause on COMP_ID and COMP_ADDRESS. What indexes on COMPANIES should I create in this case? First unique on COMP_ID, second composite on COMP_ID and COMP_ADDRESS? Or one unique index on COMP_ID and COMP_ADDRESS will be enough?
View 5 Replies View Relatedwhat is the best practice to implement in Indexing,is it global indexing or local indexing, I would like implement one of them in object that has been partitioned horizontally.i dont know exactly what to make of it.
View 9 Replies View RelatedI am exporting using query parameter. I am trying to export subset of table using rowid.
SQL> select rowid , name from tab1;
ROWID NAME
------------------ ---------------
AAAM0rAAEAAAAGMAAA sam
AAAM0rAAEAAAAGMAAB sona
AAAM0rAAEAAAAGMAAC rose
AAAM0rAAEAAAAGMAAD chris
AAAM0rAAEAAAAGMAAE san
.................. ....
.................. ....
command given as
exp sam/sam tables=tab1 file=exprwid.dmp query="where ROWID='AAAM0rAAEAAAAGMAAA'" log=log1.log
Export done in US7ASCII character set and AL16UTF16 NCHAR character set
server uses WE8ISO8859P1 character set (possible charset conversion)
About to export specified tables via Conventional Path ...
. . exporting table TAB1
EXP-00056: ORACLE error 911 encountered
ORA-00911: invalid character
Export terminated successfully with warnings.
how can i export this record ?
I'm Using Oracle DB, I have got performance issue in one of my portlet, Can Re-Indexing work for Increasing the performance.
View 3 Replies View Relatedi want to ask about indexing in partition table. i have table that indexed by local index. when i want to select all data. I execute this query
select * from Book_Issue_Part where status='Pinjam';
but it does not select all data, only partly data have selected. is it a wrong query to select all data in indexing partition table? so what query should i execute to get all data.
however im taking database size into account. i believe there has to be some sort of medium when it comes to infrastructure and performance.
this is specific to tables which map other tables. these tables that im creating usually consist of only 4 columns. 3 columns are mapped ids and the other a date in field. the 3 columns need to be indexed because they are used in joins within views. also those fields in the other tables are indexed as well.
DB: Oracle 10g R2
OS: SUSE Linux SP 1 x864
We are maintaining a DR of our Database Server(oracle 10g R2 atop SUSE SP1 Linux) using Platespin(
[URL]......
Platespin is set to replicate(block based), incremental data(delta) every 1.5 hour from Production to DR site over a 30 Mbps dedicated fiber link.
Our maximum changes of data per day(during business hours) wont exceed 300 MB. During business hours Platespin replicates at least 1 GB at every replication cycle, while during off hours it replicates 300 to 500 MB per replication cycle. We are facing this strange issue with this box only(SLES 10 SP1 + Oracle 10g R2), we have protected MS Exchange 2007 Server based workloads without this strange issue, i.e in case of Exchange only delta replicates from Production server to DR site on Platespin.
Platespin support says us that Oracle re-indexes its database for better performance, so it is possible that re-indexing causes the blocks level changes on the storage, and since Platespin works on Block level, thats why it replicates so much(even though data is not changed that much)
here is actual words of Platespin support
<snip>
I think whenever Oracle database Indexing happens, it changes almost most of the Blocks of database and Platespin replicate all those Blocks.
As you know, Platespin checks the Date/Time attribute of every blocks before replication and if Date/Time attribute changes from last replication, it considers as changed block and replicate those blocks on Platespin Appliance. So, my suggestion is just look into the Oracle server behaviour before/after Data indexing process and do needful or do some workaround to overcome this issue.
</snip>
is there anything we can do at oracle level ?
I have two databases called
a) Old Retailer database
b) New Retailer Database
We cant delete old database but we are using and we want to use both of them.
Now problem
Both dbs have a table called retailer.
I want to prevent duplication in them.
I.e it there is a retailer 12 created in old one new one shldnt allowo us to create retailer
id---12
similarly if there is a retailer 13 in new one it shouldnt be created in old table
retailer ids are unique. its not the case that with retailer id for example 5 there is a
different retailer in these two tables of diffferent DBS. Its unique
Now I have to put this criteria
how can i do it?
Second questions===================
if a primary key is auto increment. how can i skip some numbers lets say it is 1,2,3,4
i want that when next record shld be inserted it shld be 6 not 5. i dont want to feed the value 6. i want to know how to change indexing of primary key to skip some numbers when its set to auto increments.
My procedure proc_ex is in when_validate_item trigger
I have one Multi Record data block in my form with values in its items
I need to Populate multi record block values to one Record Group using
add_group_row,
add_group_column,
set_group_char_cell to populate values to record group
Let us suppose my multi record data block looks like
item1 item2 item3 item4
10 20 50 70
25 15 30 45
45 90 47 38
75 25 85 90
30 56 78 80
how to populate these multi record datablock values to Record Group ???..Eagerly waiting for your Replies
I am loading data file using SQL Loader in TOAD 10.5.1.3 in Oracle 10g using the control file below and loading data into 2 tables post1.thead and post1.tdetl. THEAD contains item level transaction and TDETL is detail level when a transaction has a discount or promo attached to for that item. When the
LOAD DATA
APPEND
INTO TABLE post1.thead
WHEN (1:5) = 'THEAD'
TRAILING NULLCOLS
(file_type POSITION(1:5)CHAR,
[code]....
A particular THEAD value may have 0, 1 or many TDETL corresponding values.below is a sample data file. When the position 21 in the TTAIL has a value of 1 or 2, then we know that there is a promo or discount applicable to the ITEM (THEAD).
THEAD0000000002201109142011091400000000000002091 1 0 -1
TTAIL0000000003000000
THEAD0000000012201109142011091400000000000002091 1 0 -1 20110914-1
TDETL0000000013CMPSPL1004 0 000000010000P0000000000000019990000000000000000200000
TTAIL0000000014000001
THEAD0000000021201109142011091400000000000002091 1 0 -1
TDETL0000000022EMPDSC1005 0 000
TDETL0000000023SCOUP 1006 0 973452 000
TTAIL0000000024000002
What I want to acheive is to accurately reflect a TDETL to its corresponding THEAD, as both THEAD and TDETL are loaded into separate tables. How can we have the 2 records correlated?
I have a table which contains the multiple records for single ID No. Now i have to select single record which contains the latest date. here is the structure Name
Null Type ------ ---- ------------ ID_P NUMBER NAME_P VARCHAR2(12) DATE_P TIMESTAMP(6) Records---------------------1 loosi 22-AUG-13 01.27.48.000000000 PM1 nammi 26-AUG-13 01.28.10.000000000 PM2 kk 22-AUG-13 01.28.26.000000000 PM2 thej 26-AUG-13 01.28.42.000000000 PM
now i have to select below 2 rows how can write select qurie for this?
1 loosi 26-AUG-13 01.27.48.000000000 PM2 thej 26-AUG-13 01.28.42.000000000 PM
I have a fairly standard Purchase Order form which contains pre-loaded data (been uploaded from an XML file).When the Purchase Order is processed, the form updates a Price History table only if the Price on the PO_Details changes.The code for updating the price history table is contained in a PRE_UPDATE trigger on the PO_Details Data Block.
No other data changes on the PO_Details table.I now want to change this so that the Price History table is updated even if the price does not change i.e I want to create a history record for each record on the PO_Details irrespective of whether it was updated or not.
Is there an alternative trigger that I can move my code to (ie move it from PRE_UPDATE) to some other trigger that is fired for each PO_Details record even if there is no change.
I have the following case to solve:
Example Table:
Nr_ordPos1Pos2Itemqty
O4018510000107 170,00
O4018520000107 30,00
O40651010000107 500,00
O40651020000107 50,00
O4114510000107 300,00
O31141010000107 50,00
O3114520000107 50,00
I need to create a query that returns record by record a field qty_progr with the cumulate qty considering previous records. The result should be the following:
Nr_ordPos1Pos2Itemqty qty_progr
O4018510000107 170,00 170,00
O4018520000107 30,00 200,00
O40651010000107 500,00 700,00
O40651020000107 50,00 750,00
O4114510000107 300,00 1050,00
O31141010000107 50,00 1100,00
O3114520000107 50,00 1150,00
I create a data block on a table when I am inserting new record only one record have been saved. Last record no longer exist.
View 4 Replies View RelatedI have one multi record block in that i have one LOV..If i choose lov in first record that won't be display in second record lov...
View 4 Replies View RelatedUsing 11gR2, windows 7 client machine. I need to update the table missing_volume (below), where I need to calculate the estimated_missing column. The calculation of estimated_missing column for current month needs previous month numbers (as commented inside the code below). I want the output like the first table. Notice the records start from January, hence estimated_missing for January can't be calculated, but for the the rest of the months it can be done by simply changing 'yr' and 'mnth' (commented inside the code towards the end).
yr mnth location volume actual_missing expected_missing estimated_missing
---------------------------------------------------------------------------------------------------------------------------------
2013 January loc1 48037 24 57
2013 February loc1 47960 3660 53 24
2013 March loc1 55007 78 57 28
2013 April loc1 54345 72 58 77The code:
UPDATE missing_volume g
[Code]....
The code does calculate correct number for 'estimated_missing' as I run the code for each month, but the problem is while updating the current month it also erases the record for previous month. E.g. as can be seen below, after I updated April the column only has the record for April, previous month record is gone, similarly updating March removed February, etc. I can't understand why it's happening!! Here is the output I get:
yr mnth location volume actual_missing expected_missing estimated_missing
---------------------------------------------------------------------------------------------------------------------------------
2013 January loc1 48037 24 57
2013 February loc1 47960 3660 53
2013 March loc1 55007 78 57
2013 April loc1 54345 72 58 77
why it's happening (I mean where is the flaw in the code) and how to get the desired output (first table).
I have a date column in a table, I need to fetch records from 27:11:11 00:00:10 to 30:11:11 00:00:10 (DD:MM:YY HH:MI:SS). i.e between seconds/min/hrs.
expecting Something like this,
SELECT * FROM TABLENAME WHERE MODEDATE BETWEEN
TO_CHAR('27/11/11 00:00:10','DD-MN-YY HH24:MI:SS') AND
TO_CHAR('30/11/11 01:10:10','DD-MN-YY HH24:MI:SS')
i want to get the last month starting from today onwards.
1) select the record of last month that is if this month is june then select the record for 1st may to 31st of may
I have a cursor which I am opening and then looping through. Within this loop I am comparing attributes within this cursor with attributes from another loop that this one is within.
you will see in the IF statements (there are several distinct IF statements within the loop) that there is a check which assesses if the attributes are not equal.
If they are NOT, the value of v_mismatch is set to 1
What I need to do instead of setting this to 1, is to go to the next record in the loop. How can I achieve this?
OPEN c_distMatrix;
LOOP
FETCH c_distMatrix INTO r_dist;
EXIT
WHEN c_distMatrix%NOTFOUND;
--compare each field and update the counter
[code].....
Based on ACCOUNT_NUMBER column we have to check the GUID data in test table.
for example ACCOUNT_NUMBER =11 as duplicate it has 2 values then only, we need to check the corresponding GUID if any occurence ("9f680174-cb87-4f71-887a-92" and "9f680174-cb87-4f71-887a-91"),
we should select,if not leave it.
CREATE TABLE test
(
GUID VARCHAR2(32 BYTE),
ACCOUNT_NUMBER NUMBER(30),
INDIVIDUAL_ID NUMBER(13)
)
[Code]...
it tried as like as below it's working.
select t2.* from (select distinct t1.guid
from test t1,( select ACCOUNT_NUMBER from test
group by ACCOUNT_NUMBER having count(ACCOUNT_NUMBER
[Code]...
I need select last added/inserted records where value = 'test1' and = 'test2' and = 'test3' in column 'NAME' should be one record for each 'test1','test2' ...
aslo..table have column 'DATE_USR' with data_type = timestamp(6)
I have a problem where my record is updated yesterday. How to get the record before updated?
View 4 Replies View Relatedin my loop with cursor in a procedure body i am displaying some field values to create a report and after every record i am displaying horizontal line(-------) but i don t want this line to appear after the last record displayed like below,
gfsfsf gsgfsfds gsdgfdg
------------------------------------
edyet gdgtdgt gtdfdfdgd
------------------------------------
dfds hedhgg idudhdh
how can i achieve this within pls/sql procedure body.
restricting a subquery's results to the first record. If I use the following:
SELECT WU.DISCREP, M.PART, M.ETA
FROM DB.MICAP M
LEFT JOIN DB.WRITEUPS WU ON M.WRITEUPID=WU.WRITEUPID
LEFT JOIN DB.WUC WUC ON WU.WUCID=WUC.WUCID
WHERE (WU.AIRCRAFTID=205) AND (WU.CORRECTED=0)
ORDER BY WU.PACER DESC, WUC.WUCCODE
I get
-------------------------------------------------------
#5 GEN KICKED OFFLINECABLE02-MAR-08
#5 IDG HARNESS REQUIRES WIREWIRE24-MAY-08
#1 GEN WENT OFFLINECABLE02-MAR-08
AERIAL REFUEL DOORS W/N OPENLINEAR DEL29-SEP-07
AERIAL REFUEL DOORS W/N OPENCHECK VLV30-SEP-07
WEATHER RADAR TILT KNOB HASRADAR COMP05-OCT-07
But I need the highlighted line eliminated. I've tried DISTINCT subqueries in the WHERE M.WRITEUPID IN (SELECT DISTINCT... various experiments with joins, etc.
I have an alerts system for a web site built in an Oracle DB. There are multiple alerts that can be assigned per page, but I want to display only the most recent. I have 5 pages but each page can have different alerts.
I have a parent table of all the alerts and a child table with the page assignments. I want to pull one record (the newest) for each page assignment. The pages can increase so I can't hardcode this.
I keep thinking I need to do something like:
SELECTalertID, headline
FROMalerts_V
WHEREalertID IN (
SELECTdeptID
FROMso_departments
)
But it's obviously not quite what I need and not the correct syntax.
I am unable to select only record for 07Nov06. if i use between i get records e.g. :07 November ,2006-0941
Select
TO_CHAR(session_START_time,'dd month,yyyy-hh24mi') LOG_IN_TIME,
TO_CHAR(session_END_time,'dd month,yyyy-hh24mi') LOG_OUT_TIME
from SESSION_LOG
where SESSION_LOG.SESSION_START_TIME
between '06Nov06' AND '07Nov06';